https://gcc.gnu.org/g:68316afcb5028bc127f82f26dcaae5b0d3846a6c
commit r14-10772-g68316afcb5028bc127f82f26dcaae5b0d3846a6c
Author: Jakub Jelinek <ja...@redhat.com>
Date: Fri Oct 4 13:12:45 2024 +0200
i386: Fix up ix86_expand_int_compare with TImode comparisons of SUBREGs
from V8{H,B}Fmode against zero [PR116921]
The following testcase ICEs, because the ix86_expand_int_compare
optimization to use {,v}ptest assumes there are instructions for all
16-byte vector modes. That isn't the case, we only have one for
V16QI, V8HI, V4SI, V2DI, V1TI, V4SF and V2DF, not for
V8HF nor V8BF.
The following patch fixes that by using the V8HI instruction instead
for those 2 modes. tmp can't be a SUBREG, because it is SUBREG_REG
of another SUBREG, so we don't need to worry about gen_lowpart
failing.
2024-10-04 Jakub Jelinek <ja...@redhat.com>
PR target/116921
* config/i386/i386-expand.cc (ix86_expand_int_compare): Add a SUBREG
to V8HImode from V8HFmode or V8BFmode before generating a ptest.
* gcc.target/i386/pr116921.c: New test.
(cherry picked from commit 92e9e971ced90af5a825ae4b35ad6c98c9ab86da)
Diff:
---
gcc/config/i386/i386-expand.cc | 2 ++
gcc/testsuite/gcc.target/i386/pr116921.c | 12 ++++++++++++
2 files changed, 14 insertions(+)
diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
index 51efe6fdd7d0..1a15388d9099 100644
--- a/gcc/config/i386/i386-expand.cc
+++ b/gcc/config/i386/i386-expand.cc
@@ -3063,6 +3063,8 @@ ix86_expand_int_compare (enum rtx_code code, rtx op0, rtx
op1)
&& GET_MODE_SIZE (GET_MODE (SUBREG_REG (op0))) == 16)
{
tmp = SUBREG_REG (op0);
+ if (GET_MODE (tmp) == V8HFmode || GET_MODE (tmp) == V8BFmode)
+ tmp = gen_lowpart (V8HImode, tmp);
tmp = gen_rtx_UNSPEC (CCZmode, gen_rtvec (2, tmp, tmp), UNSPEC_PTEST);
}
else
diff --git a/gcc/testsuite/gcc.target/i386/pr116921.c
b/gcc/testsuite/gcc.target/i386/pr116921.c
new file mode 100644
index 000000000000..8cd4cf93b76d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr116921.c
@@ -0,0 +1,12 @@
+/* PR target/116921 */
+/* { dg-do compile { target int128 } } */
+/* { dg-options "-O2 -msse4" } */
+
+long x;
+_Float16 __attribute__((__vector_size__ (16))) f;
+
+void
+foo (void)
+{
+ x -= !(__int128) (f / 2);
+}
